WASHINGTON — Browsing for the best deal on the Internet during this holiday season can make even the savviest shopper feel like a high-stakes gambler.
Online prices for everything from food processors to flat-screen TVs can fluctuate by hundreds of dollars, from site to site, day to day, or even hour to hour. For bargain-hunting Web surfers, such volatility presents a vexing dilemma: Click buy now, only to see the product selling for less the next time you log on, or hold off and risk the price jumping out of reach.
A Nikon D7000 digital camera and lens, for example, start at a list price of $1,496.96. After Best Buy published a Black Friday teaser price of $999, other retailers slashed prices online.
At 12:58 a.m. EST on Thanksgiving, CompUSA offered the camera for $996.95, according to the price-tracking website Decide.com. Seventy minutes later, Walmart dropped its price to $999. Within the hour, Target followed suit, and before dawn Friday, competitors Amazon, B&H and other sellers had done the same.
The following week, the Nikon D7000 jumped to a range of $1,196.95 to $1,396.95 on those same websites.
Its always been true let the buyer beware, and now we can say thats also true online, said Jeffrey Chester, the executive director of the nonprofit Center for Digital Democracy in Washington. Todays better deal may be tomorrows raw deal.
Price wars rage with particular intensity during the holidays, but skirmishes go on year-round. Prices rise as often as they fall, and its all but impossible for shoppers to discern any method in the madness.
The bottom line is this is so confusing and so data intensive that its a job for a machine, said Oren Etzioni, a computer science professor at the University of Washington in Seattle.
Online price changes are driven by software that enables retailers to react to competitors in real time and make personalized pitches to consumers based on their browsing and purchase histories.
Complex algorithms calculate factors such as inventory levels, how many sellers are carrying a product, how long its been on the market and when a new model is expected. Retailers also use browser data to target customers with discounts or free-shipping offers for products they might like depending on where they live, their demographic profiles, the pages theyve clicked on, the items in their baskets, even the local weather.
The information used to customize your online shopping experience is culled from your click history and cookies, tiny data files saved on a computers hard drive while visiting a website. If youre a repeat customer or a member of a loyalty program, marketers know even more about your tastes. The goal is to turn the stores website or app into a virtual personal shopper for each visitor.
Retailers have a much better idea of who you are and what youre willing to buy, said Jason Buechel, a senior executive at Accenture, a technology outsourcing and consulting company. Instead of having you price-compare across multiple websites, theyre trying to drive you to close the sale as soon as possible.
Does that mean retailers might offer you a different price from another customer for the same product? Its hard to say. Although the technology exists, industry professionals say its not a widely accepted practice, mainly because companies fear a backlash.
Amazon took a public relations hit in 2000 when customers noticed that the company had sold the same DVD for different prices. Once the story broke, the company said the phenomenon was part of a temporary pricing experiment, and it refunded customers.
More recently, Orbitz came under fire for showing pricier hotels at the top of search results to Macintosh users, who have higher household incomes than PC users do, research shows. As Chief Executive Officer Barney Harford explained in a May blog post, the intent was to give consumers more personalized recommendations.
Weve identified that Mac users are 40% more likely to book a four- or five-star hotel than PC users, Harford wrote. A similar skew applies for iPad users. We can use that information to influence which hotels we recommend to users we see searching on a Mac or an iPad versus a PC, for example.
Retailers often use the Web as a laboratory to test pricing strategies, said Joe Stanhope, a customer intelligence analyst for Forrester Research Inc.Sometimes its as simple as trying to determine if people prefer a certain dollar amount off, or a certain percentage off, and which type of offer resonates better with customers, Stanhope said. Theyre trying to understand what the real parameters are in pricing. The reason they like to do it on the Web is you have a lot of traffic and a lot of people buying things.
The lack of transparency troubles consumer advocates.
Your financial future is increasingly going to be determined by these secret algorithms powered by computers, said Chester, from the Center for Digital Democracy. All thats happening under the radar or behind the curtain in a digital wizards lab, and I think its incredibly unfair.
The good news is that consumers have a growing number of weapons in their digital arsenal.
The proliferation of mobile devices has made comparing prices and discount shopping easier than ever, thanks to websites and apps such as Bizrate, PriceGrabber and CouponCabin.
Decide, the cutting-edge site founded last year by Etzioni, the University of Washington professor, helps cost-conscious consumers make sense of price changes using the same data-driven analytics that retailers use to predict consumer behavior.
The site analyzes millions of reviews, blogs, articles, news releases and pricing trends, then recommends whether to buy or wait. In the case of the Nikon camera, Decide this week recommended waiting, since, it says, the price is likely to drop $158 or hold steady.
The predictions arent perfect, but the site claims theyre right 77 percent of the time.
Often consumers are the target of big data and of data mining, and what were doing is were leveling the playing field, Etzioni said. Were trying to make data be your friend rather than a weapon thats used against you.
Email: email@example.com; Twitter: @lindsaywise